Kinematics of a total arthroplasty of the ankle: comparison to normal ankle motion

Although the concept of a total ankle arthroplasty has been advanced as a method for treating severe ankle arthritis, the clinical experience with all of the models developed has been discouraging. Both the constrained designs, which maximize joint contact area by restricting the available motion, and the unconstrained designs, which allow more normal motion at the expense of higher contact stresses, uniformly result in implant loosening, pain, and clinical failure in 2 to 7 years. This has led to the recommendation against the use of a total ankle arthroplasty except in very low-demand patients. Failure of ankle implants can be ascribed to either anatomic considerations (e.g.--the talus is too small to accommodate the stress transfers of a prosthesis), or mechanical etiologies. Abnormal 3-dimensional motion of the ankle following arthroplasty would fall into the latter category. This study examined the motion that occurs after implantation of an unconstrained-type total ankle arthroplasty. Using previously validated methodology, axially loaded ankle specimens were cycled through an arc of plantarflex/dorsiflexion while measuring the resulting coupled internal/external and varus/valgus rotations. The average coupled motions in prosthetic ankles were not significantly different than their intact controls. There was, however, a significantly increased amount of hysteresis (defined as the difference between the upper and lower pathways of coupled motion at any given sagittal position) that occurred as the ankle was dorsiflexed and plantar flexed. The increased hysteresis was seen in both the axial and coronal planes. This indicates that there was a greater permitted envelop of motion in the prosthetic ankles compared to normal ankles. It is hypothesized that this subtle change in ankle kinematics caused by the arthroplasty leads to abnormal stress transfer at the prosthesis-bone interface, thereby promoting early implant failure.